CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Methods: Child anthropometric parameters were measured six-monthly using standardized procedures in the ongoing PROMOTE observational study of women with HIV and their children. Enrolment occurred between December 2016 and June 2017. The WHO child-growth standards (2006) were used to calculate age-and-sex appropriate Z-scores for weight (WAZ), height (HAZ), and weight-for-height (WHZ). Severe growth-faltering (stunting, underweight and wasting) was defined as more than two standard deviations below the WHO population median, respectively. Generalised estimating equations (GEE) were used to assess correlates of stunting including maternal factors (age, education), country, infant sex, and surrogate measures of household level sanitation and socioeconomic status (tap water use, size of the house). Results: Of the 1459 HEUs aged 2-5 years during the study period included in this analysis, 48.5%were female. Mean (sd) z-scores were below population norms for height (-1.2 +1.2) and weight (-0.5 +1.0) across all 4094 repeated measurements; 934 (22.9%) were stunted, 208 (5.1%) underweight and 72 (1.8%) wasted. We found that Malawi location when compared to South Africa [adjusted odds ratio; (95% CI): 2.50; 1.74-3.60] and being born to a mother who did not complete secondary school (1.47; 1.11-1.95) were associated with higher odds of stunting; whereas older children had lower odds of stunting (0.96; 0.95-0.96). Conclusion: High rates of growth faltering were observed in this large multi- country cohort of predominantly breastfed African children who survived to at least 2 years and escaped HIV infection. Early interventions are necessary to address malnutrition in the growing population of HEUs in order to optimize their health and future human capital. Maternal factors, specifically education may be a key area of focus.

obese women and similar among HIV-, not overweight and HIV+, overweight/ obese women. At 12 months postpartum, the probability of infant overweight was highest for infants born to HIV-, overweight/obese women (0.24;95%CI 0.19, 0.30), followed by HIV+, overweight/obese women (0.19; 95%CI 0.13; 0.24), HIV-, not overweight women (0.13; 95% CI 0.07, 0.18), and lowest among HIV+ not overweight women (0.10; 95%CI 0.05, 0.15). Conclusion: In this setting of high maternal BMI, HEU infants had lower mean BMI during the first year of life than HU infants, regardless of maternal BMI status. By 12 months postpartum, the probability of being overweight was >10% for all groups. However, infants born to overweight/obese women, compared to not overweight women, were more likely to be overweight, regardless of HIV status. Future research should examine if maternal HIV and BMI status increase the risk of cardio-metabolic complications for HEU infants later in life.

Poster Abstracts

802 DISTINCT CORD C-PEPTIDE, ADIPOKINE, AND LIPIDOMIC SIGNATURES BY IN UTERO HIV EXPOSURE Jennifer Jao 1 , Lauren Balmert 1 , Shan Sun 2 , Thomas Kraus 3 , Brian Kirmse 4 , Mitchell Geffner 5 , Yunping Qiu 6 , Stephen M.Arpadi 7 , Elaine J.Abrams 8 , Derek LeRoith 3 , Rhoda Sperling 3 , Irwin J. Kurland 6 1 Northwestern University, Chicago, IL, USA, 2 Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, USA, 3 Icahn School of Medicine at Mt Sinai, New York, NY, USA, 4 University of Mississippi Medical Center, Jackson, MS, USA, 5 University of Southern California, Los Angeles, CA, USA, 6 Albert Einstein College of Medicine, Bronx, NY, USA, 7 Columbia University, New York, NY, USA, 8 ICAP at Columbia University, New York, NY, USA Background: Metabolic derangements early in life of HIV-exposed uninfected (HEU) infants have been reported. Methods: Pregnant HIV+ and HIV- women were enrolled with their infants in a US cohort from 2009-15. We measured insulin, C-peptide, and adipokines [metabolic (resistin, leptin) and inflammatory (Interleukin (IL)-6, Tumor Necrosis Factor-a (TNFa)] in cord blood of HEU and HIV-unexposed uninfected (HUU) infants using multiplex ELISA. Demographic, clinical, and in utero antiretroviral therapy (ART) exposure data were collected. Metabolites and lipid subspecies were measured via mass spectrometry. Linear regression models were fit to assess the association of in utero HIV exposure with cord insulin and C-peptide. Orthogonal partial least squares discriminant analysis (OPLS-DA) was used to assess if differences in metabolites and lipid subspecies discriminate between HEU and HUU infants. Elastic net regression was used to identify factors including metabolites and lipid subspecies most associated with increased cord C-peptide, stratified by in utero HIV exposure. Results: Of 118 infants, 56 were HEU. No differences in maternal race/ethnicity, pre-pregnancy BMI, gestational diabetes (GDM) or infant preterm birth (PTB), birth weight/length were noted. All HEU were ART-exposed (52% PI, 21% NNRTI, 9% INSTI). After adjusting for maternal age, GDM, family diabetes history, pre-pregnancy BMI, as well as infant sex, PTB, and birth weight z score, mean cord insulin (ß=0.295, p=0.03) and C-peptide (ß=0.522, p<0.01) were significantly higher in HEU vs. HUU infants. IL-6 correlated positively with C-peptide in HEU (rho=0.30, p=0.05) but not HUU infants (rho=0.08, p=0.52) while resistin correlated inversely with C-peptide in HUU (rho= -0.40, p<0.01) but not HEU. Leptin correlated positively with C-peptide in both groups (rho=0.64, p<0.01 in HEU; rho=0.26, p=0.04 in HUU). OPLS-DA showed clear group separation by metabolites and lipid subspecies. (Fig)

801 HIGH BODY MASS IN HIV+ & HIV– WOMEN AND THEIR HIV-UNINFECTED INFANTS IN SOUTH AFRICA Angela Bengtson 1, Stanzi M. le Roux 2 , Tamsin K. Phillips 2 , Kirsty Brittain 2 , Allison Zerbe 3 , Thoko Malaba 2 , Greg Petro 2 , Hlengiwe Madlala 2 , Elaine J. Abrams 3 , Landon Myer 2 1 Brown University, Providence, RI, USA, 2 University of Cape Town, Cape Town, South Africa, 3 Columbia University, New York, NY, USA Background: HIV-exposed uninfected (HEU) infants may have altered growth relative to HIV-unexposed (HU) infants. Maternal body mass index (BMI) and obesity are strongly linked to child growth, but there are few data on associations between maternal BMI and HEU infant growth. Methods: We followed cohorts of HIV+ (initiating TDF+XTC+EFV) and HIV- women from first antenatal visit (ANC) through 12 months postpartum with their breastfed infants. We estimated pre-pregnancy maternal BMI (kg/ m 2 ) from routine antepartummeasures; trained staff collected postpartum anthropometry (maternal BMI and infant BMI Z-scores). Virological testing excluded infant HIV infection. We examined relationships between a combined exposure of maternal HIV/ART (HIV+ vs HIV-) and pre-pregnancy BMI (not overweight (BMI<25) vs overweight/obese (BMI >25)) with mean infant BMIZ scores using multivariable linear mixed models and the probability of infant overweight (BMIZ >2 SD) using modified Poisson regression. Results: In 780 mother-infant pairs (49% HIV+mothers), 68% had pre- pregnancy BMI ≥25; 4%were underweight (BMI<18.5). HIV+ women were less likely to be overweight/obese (65% vs 71%), but more likely to report alcohol use (25% vs 7%) and food insecurity (17% vs 3%) vs HIV- women. HIV- women were more likely to breastfeed for ≥6 months (62% vs 44%); breastfeeding ≥6 months did not differ by BMI status (~54%). Throughout follow-up, infant BMIZ trajectories differed by maternal HIV-BMI status (Figure): scores were lowest among HIV+, not overweight women and highest among HIV-, overweight/

CROI 2020 297